# Exploring Tradeoffs in Models for Low-latency Speech Enhancement

arXiv: https://arxiv.org/abs/1811.07030 · HTML: https://mlsanity.com/paper/1811.07030/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Nov 2018.

## Built on by (most cited first, 8 of 8)

- [CMGAN: Conformer-based Metric GAN for Speech Enhancement](https://mlsanity.com/paper/2203.15149/) (2022)
- [CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement](https://mlsanity.com/paper/2209.11112/) (2022)
- [Real-time Denoising and Dereverberation with Tiny Recurrent U-Net](https://mlsanity.com/paper/2102.03207/) (2021)
- [Dynamic nsNet2: Efficient Deep Noise Suppression with Early Exiting](https://mlsanity.com/paper/2308.16678/) (2023)
- [Real-Time Audio-Visual Speech Enhancement Using Pre-trained Visual Representations](https://mlsanity.com/paper/2507.21448/) (2025)
- [Short-Term Memory Convolutions](https://mlsanity.com/paper/2302.04331/) (2023)
- [Towards Sub-millisecond Latency Real-Time Speech Enhancement Models on Hearables](https://mlsanity.com/paper/2409.18239/) (2024)
- [Diffusion Buffer for Online Generative Speech Enhancement](https://mlsanity.com/paper/2510.18744/) (2025)

## Abstract (arXiv)

We explore a variety of neural networks configurations for one- and two-channel spectrogram-mask-based speech enhancement. Our best model improves on previous state-of-the-art performance on the CHiME2 speech enhancement task by 0.4 decibels in signal-to-distortion ratio (SDR). We examine trade-offs such as non-causal look-ahead, computation, and parameter count versus enhancement performance and find that zero-look-ahead models can achieve, on average, within 0.03 dB SDR of our best bidirectional model. Further, we find that 200 milliseconds of look-ahead is sufficient to achieve equivalent performance to our best bidirectional model.